Toward Efficient and Robust Multiple Camera Visual-inertial Odometry

TL;DR

This paper introduces a GPU-enhanced multi-camera visual-inertial odometry system that improves efficiency and robustness, reducing CPU load and latency while increasing stability and accuracy in pose estimation.

Contribution

It presents a novel front-end using NVIDIA VPI for multi-camera VIO, significantly enhancing efficiency and robustness over existing methods.

Findings

CPU resource occupation reduced by 40.4%

Computational latency decreased by 50.6%

Higher VIO initialization success rate and improved robustness

Abstract

Efficiency and robustness are the essential criteria for the visual-inertial odometry (VIO) system. To process massive visual data, the high cost on CPU resources and computation latency limits VIO's possibility in integration with other applications. Recently, the powerful embedded GPUs have great potentials to improve the front-end image processing capability. Meanwhile, multi-camera systems can increase the visual constraints for back-end optimization. Inspired by these insights, we incorporate the GPU-enhanced algorithms in the field of VIO and thus propose a new front-end with NVIDIA Vision Programming Interface (VPI). This new front-end then enables multi-camera VIO feature association and provides more stable back-end pose optimization. Experiments with our new front-end on monocular datasets show the CPU resource occupation rate and computational latency are reduced by 40.4% and 50.6% without losing accuracy compared with the original VIO. The multi-camera system shows a higher VIO initialization success rate and better robustness overall state estimation.